Gendered political space: civil society, contingency theory, and the substantive representation of women

ABSTRACT Whilst existing civil society studies generally fail to systematically examine the way that contextual factors shape women’s representation in the civil sphere, political science has predominantly focused on legislative settings. This article responds to the resultant knowledge-gap by examining the hitherto underexplored role of civil society as a political space integral to the substantive representation of women (SRW)—or, the process by which women’s concerns are advanced in policy and politics. The article uses grounded theory in order propose a systematic analytical model showing how the SRW is a contingent process whereby the motives of civil society organizations are translated into action repertoires shaped by three (non-discrete) spheres: political, socioeconomic, and organizational. Its wider contribution to civil society scholarship is in highlighting how civil society is a complex, heterogeneous political space wherein SRW claims-making requires cognizance of the co-presence of contingent factors that offer immanent explanatory power.


Introduction
As a leading account highlights, 'it is curious that there has been so little interrogation of the relationship between gender and civil society within either feminist or civil society theories. This is surprising…because each set of theories would have much to gain from the other in terms of theorization and practical knowledge' (Howell, 2005, p. 28). In a similar vein, earlier analysis of women's political empowerment concluded: 'the terms 'state versus market' are well known and frequently used as if the distinction was clear-cut. However, civil society is often left out of the discussion…civil society involves gender conflict…To leave out any part of the triangle, obscures the debate' (Dahlerup, 1994, pp. 117-119). A survey of the extant literature suggests limited progress in the two decades since the foregoing observation was made. Thus, for example, Kang (2014, p. 86, emphasis added) concludes: 'civil society has advocated for the representation of women's interests. Yet, relatively little is known about the full range of actors who seek the representation of women's interests, mobilize around women's issues, and articulate specific preferences…Who in civil society seeks to influence the representation of women's interests and how'? This article offers a starting point to address these lacunae. Specifically, it presents an analytical model for exploring civil society's role as a political space integral to the substantive representation of women (SRW).
Addressing this knowledge-gap is important for, as Annesley (2010) notes, whilst much of the existing political science literature has emphasized the contingent nature of the SRW, it has nevertheless focused on legislative settings. Yet this offers only a partial view: effective representation cannot be divorced from context, and analysis of the SRW needs to be cognizant of the multiple, overlapping arenas in which it takes place (Chaney, 2014;Eto, 2012). As Marx, Ferree, Gamson, Gerhards, and Rucht (2002, p. 307) note: there is a need to examine the practices of power diffused outside formal political institutions … We call this tradition constructionist because it emphasizes the contingently produced nature of every aspect of the political process … [there is need to … ] begin from feminist premises and develop their theories in part to explain and critique the marginality of women in politics.
In taking up this call, scholarly analysis therefore needs to pay systematic attention to the manifold and contingent ways civil society organizations' (CSO) motives are translated into actions. Accordingly, the following discussion uses grounded theory from a survey of the political science literature and empirical data in order propose a systematic analytical framework of women's representation.
In definitional terms, the SRW refers to the situation whereby women's needs and concerns are reflected in public policy and politics. In turn, the present aim is twofold: as noted, to further understanding of the nexus between civil society and women's representation, and-by producing an analytical framework-to provide a resource for future empirical investigation and theory-building.
Civil society is an appropriate focus to further contemporary understanding of women's representation because CSOs are positioned at a key social and political juncture as they simultaneously seek to exercise 'mandates to advocate, politicize and provide services for specific groups of women' (George, 2007, p. 682). From a global perspective, and as a range of studies attests (cf. Bee & Guerrina, 2014;Jones & Kas, 2005;Sener, 2014), such activities are consonant with governments' espousal of gender equality and representative democracy. They also feature prominently in a series of international treaties, human rights instruments, and equalities law. Collectively, these underline the right of women from across civil society to engage in politics and policy-making (Kabeer, 2005;Pascall & Lewis, 2004).
In order to address the study aims, the remainder of this article is structured thus: a summary of the contribution of this article to civil society studies and a discussion of civil society as a gendered political space is followed by an overview of the SRW. After the methodology, contingency theory is discussed and the current study's analytical framework is outlined using grounded theory in order to show how the translation of CSO motives into actions is shaped by three spheres: the political, socioeconomic, and organizational.
presented theory-based analytical model of the SRW add to academic work on the nature and role of civil society? As the following reveals, the answers come from each phase in the civil society 'metanarrative' (cf. Powell, 2007, p. 37).
In classical scholarship the Aristotelian view emphasized that civil society is indistinguishable from the state. In the Polis (or city state), both spheres are merged, constituting an 'association of associations' (Pérez-Díaz, 2014, p. 819). This constellation enabled citizens to share in the virtuous task of ruling and being ruled. As Edwards (2009, p. 6) notes, according to this conception 'the state described the "civil" form of society-and "civility" described the requirements of good citizenship'. The latter resonates with contemporary debates on gender relations and citizenship. Thus, for example, Lister (2012, p. 84) reflects on the way that the gendered and androcentric nature of citizenship has continued through to the twenty-first century. She observes: for much of its history, a veil of gender-neutrality has obscured the nature of this differential relationship. Today, as feminist theorists have stripped away this veil, the challenge is to reconceptualize citizenship in gendered terms in the image of women as well as men.
As a burgeoning international literature attests (cf. Cheriet, 1996;Howell, 2008;Salmenniemi, 2005;Shepherd, 2015), this is a global challenge and civil society is the nexus between the state and the individual wherein gender relations and prevailing notions of citizenship are (re-)defined. It is for this reason that the present article's focus on the SRW is apposite and makes a needed contribution.
A further rationale for this article's focus stems from Enlightenment thinking: specifically, what liberal theorists such as Rousseau and Locke highlight as civil society's role in providing a defence against absolute and oppressive rule and the unwanted (and unwarranted) intrusion by the state and those in positions of power on the rights of the individual. A key point here is the need to understand how women's representation operates in civil society in ways that uphold existing gender equality rights. These stem from a diverse body of law and treaty agreements. International examples include: the United Nations' (UN) Convention on the Elimination of Discrimination Against Women, the Beijing Declaration and Platform for Action (see Chaney, 2015), UN resolutions SCR1325, SCR1888, and SCR1889; and European Commission Directives 2000/43/EC, 2006/54/ EC, and 76/207/EEC. Examples of domestic enactments include: the Equality Act (2010) in the UK, the Civil Rights Act (1991) in the USA, the Sex Discrimination Act (1984) in Australia, and Promotion of Equality and Prevention of Unfair Discrimination Act (2000) in South Africa.
In addition, post-Enlightenment thinking on civil society offers three further, fundamental reasons as to why the present focus makes a useful addition to civil society scholarship.
(1) Putnam's seminal work on social capital (inter alia, norms and networks of trust and reciprocity that link individuals) highlights the connections between associative life and civil society (Putnam, Leonardi, & Nanetti, 1993). As work by scholars such as Lowndes (2004) emphasizes, these patterns and processes of sociability are gendered in nature. Accordingly, we need to better understand how civil society operates as a political arena in which women mobilize and advance claims on state power elites.
(2) In turn, such claims-making is driven by the gendered nature of public policy and welfare (Bambra, 2004). This is particularly salient in the first decades of the twenty-first century owing to the rise of 'welfare pluralism'-or, the 'mixed economy of welfare' (Beresford & Croft, 1983;Chaney & Wincott, 2014). This is the situation whereby CSOs are increasingly drawn into shaping public policy and welfare service delivery. It is also driven by a need to challenge oppressive and discriminatory gender norms (Katuna & Holzer, 2015).
(3) The literature on feminist institutionalism is a further imperative underpinning the need for the present work (cf. Kenny, 2007). This relates to how contemporary institutions remain highly gendered in terms of access to power and resources. The organizations in civil society are no exception. For example, in general, men predominate in leadership and positions of power in many NGOs.

Civil society-gendered political space
Both the broader feminist political science literature and scholarly work on the SRW have attempted to address extra-parliamentary representation. Yet, each offers an incomplete picture. In part, this is because both strands draw selectively on elements of the same literature (e.g. Gamson, 1975;Tilly, 1995) and adapt shared concepts (e.g. issue framing, opportunity structures) (see e.g. Ferree, 2003;Lombardo, Meier, & Verloo, 2009;McBride & Mazur, 2010). The most insightful aspect of the existing work is that on 'state feminism'. As Kantola and Squires (2012, p. 382) observe, this refers to 'the alliances between women's policy agencies and women's movement activists, and their effectiveness in getting state responses to the movement's demands'. In the same vein, McBride and Mazur (2010, p. 254) describe it as an interaction founded on claims-making designed 'to produce feminist outcomes in either policy processes or societal impact or both' ('claims-making' here can broadly be defined as advancing policy demands on those in power). However, notwithstanding its utility, this strand of academic work has limitations. This is because, as Kantola and Squires (2012, p. 382) note, 'state feminism traditionally focuses on the dynamic between gender equality advocates and the state, framed by an assumption of a modernist bureaucratic state and a cohesive national women's movement'. They continue, 'in the context of governing styles in which boundaries between sectors have become blurred … state feminist analyses increasingly need to be refined'. Waylen (2008, p. 115) concurs with this. She notes that, 'despite its large remit, gendered perspectives have been almost entirely absent from any aspect of this growing political science literature on governance'. In particular, civil society and political science research have been slow to recognize the implications of revised governance practices for the SRW. These lacunae underline that the civil sphere is an appropriate locus of enquiry not only because of the rise of new forms of governance (inter alia, co-production, social enterprise, and service delivery by third sector organizations)-but also because it is seen as a core factor in maintaining the health of democracy.
The latter is predicted by pluralism (Dahl, 1961). Here exogenous civil society interests perform a pivotal role through critical engagement with the institutions of the state, shaping policy-making as part of the wider process of holding government to account.
It is not just political science that has fallen short on civil society and the SRW; welfare state theory has also struggled to come to terms with new governance practices. It has often categorized states in terms of the pattern of interplay between social policies and the structure of the labour market (cf. Esping Andersen, 1990)-or, it has offered an exclusive focus on state policy (Fraser, 1997). In both cases, it has failed to fully recognize the influence of civil society and the advancement of the SRW in order to engender social policy and welfare.
Any attempt to address the foregoing lacunae demands a nuanced view. Specifically, from a gender perspective, the concept of civil society (Cohen & Arato, 1994;Gramsci, 1971) needs to be framed in relation to the family-as well as private, civic, and political associations; and state political institutions whilst noting 'these elements are intertwined such that their boundaries are effectively seamless' (Eto, 2012, p. 78). So conceived, civil society comprises a diverse range of associational activities extending beyond the family, to encompass non-governmental organizations (NGOs), pressure groups, charities, community groups, social movements, and campaigning organizations (Keane, 1988). In short, these are the organizations related to the principal collective signifiers associated with non-government advocacy and service organizations, namely 'voluntarism', 'voluntary sector', 'third sector', 'civil society', and 'non-profit sector' (Casey, 2004;Salamon & Anheier, 1992).
The pivotal role of civil society to the SRW can be summarized thus: (1) It has been regarded as a democratic counter-balance to the power of the state (Cohen & Arato, 1994). This view is particularly germane owing to the largely male-dominated character of state institutions before and during the modern era.
(2) It is also an arena for pluralism and solidarity around gender and identity-and normative notions of equality and rights (Alexander, 1998)-as well as a place for resisting undesired aspects of capitalism and market practices (Hardt & Negri, 2000). (3) It is a normative space for civility as well (Alexander, 2005, p. 652). Civility denotes a collective disposition to attitudes and beliefs of a shared, universal notion of common humanity founded on rights and recognition (Rucht, 2009). This stems from a range of classical sources, including Hegel's invocation for 'recognition of the other' (a call addressed in Taylor's 'The politics of recognition' (1992), as well as recent works by Honneth (2005)).
In gender terms, the foregoing strands of thought resonate with Fraser's (1997) seminal work identifying recognition as a precondition of equality. It also marks civil society out as a key arena for SRW claims-making-for, in invoking a normative vision of a civil society founded on the notion of a shared humanity, it requires due attention to be paid to the historical legacy of marginalization and oppression experienced by women. Notably, this is spelt out in Young's (1990) warning that an ill-defined notion of a shared universal humanity allows privileged groups to ignore their own group specificity. In male-dominated societies, this may signal a failure to consider gender equality on the part of male political elites. Worse, it may be misused by more powerful groups to require assimilation and the modification of claims by members of oppressed groups to fit with the dominant elite's notion of 'universal humanity'. Thus Young's 'politics of difference' resonates with civil society as a setting for the SRW when she asserts that groups traditionally subject to inequality and discrimination-such as women-require group-specific action to address historical injustice and oppression-or, as she puts it: 'participating in determining one's actions and the conditions of one's action' (Young, 2000, p. 37).

The substantive representation of women
Subject to a rich and burgeoning literature, as noted, this term has traditionally described the situation whereby politics enables women's needs and concerns to be reflected in public policy-making and law (Pitkin, 1972). Of late, there has been growing acknowledgement that political science has given disproportionate attention to SRW in legislative settings and, in response, there is a need for analytical refocusing. For example, Celis, Childs, Kantola, and Krook (2014, p. 152) observe, 'substantive representation is better conceived of as a process, involving debate, deliberation, and contestation over group interests, occurring inside and outside formal institutions' (see also Childs, Webb, & Marthaler, 2010). Earlier studies have also alluded to women's representation as being contingent in nature (see e.g. Childs & Krook, 2009, p. 526;Dovi, 2002, p. 733;Mansbridge, 1992, p. 630;and Saward, 2006, p. 297). Thus, for example, Celis et al. (2014, p. 152) characterize it as 'an active, multifaceted, and contingent process, driven by a broad swathe of actors with various views on group issues and interests'. As Saward (2010, p. 36) notes, rather than being solely concerned with formal representative structures and parliamentary practices, it is 'an ongoing process of making and receiving, accepting and rejecting claims-in, between, and outside electoral cycles' (Saward, 2010, p. 36). This broader view of claimsmaking encompasses a full range of measures to influence the political and policy agenda including: protest, campaigning, boycotts, lobbying, petitioning, and policy consultation responses. Accordingly, the purpose of this discussion is to draw upon contingency theory to address the existing knowledge-gap in relation to CSOs and the SRW.

Method
The present analysis uses grounded theory to produce an analytical framework of how civil society links to the SRW. As Strauss and Corbin (1994, p. 237) explain, grounded theory is a 'methodology for developing theory that is grounded in data systematically gathered and analysed. Theory evolves during the actual research, and it does this through continuous interplay between analysis and data collection'. Accordingly, this approach is at once interpretivist and constructivist in nature (Edwards & Skinners, 2009). Its goal is to 'find meaning in an action, or to say one understands what a particular action means, [this] require[es] that one interprets in a particular way what the actions are doing' (Denzin & Lincoln, 2003, p. 296). Charmaz underlines the utility of this strategy (2003, pp. 272-273) stating that: constructivist grounded theory: remains realist because it addresses human realities and assumes the existence of real worlds … the constructivist approach assumes what we take as real, as objective knowledge and truth, is based upon our perspective … thus the grounded theorist constructs an image of reality … that is, objective, true, and external.
The present study employs theoretical sampling of 50 scholarly research studies published in monographs and peer-reviewed journals (the References include only those cited in the article). Close reading and textual analysis of the studies was accompanied by on screen manipulation of text, coding of key themes, and identification of key quotations. This was done using a basic qualitative coding software. The theoretical sample was purposive in nature; in other words, cases were selected to reflect geographical diversity and a spread of work on women's representation in civil society contexts. As Strauss and Corbin (1998, p. 202) explain, such an approach to data selection: evolves during the [research] process. It is based on concepts that emerged from analysis and that appear to have relevance to the evolving theory … the aim of theoretical sampling is to maximize opportunities to compare events, incidents, or happenings to determine how a category varies in terms of its properties and dimensions.
Nexus: contingency theory-and its application to civil society and the SRW Weighed against earlier claims that the SRW is a contingent process (Childs & Krook, 2009;Dovi, 2002;Mansbridge, 1992;Saward, 2006), the aim of the remainder of this article is to take systematic stock of factors that might support such assertions based on a survey of empirical studies, thereby producing an analytical framework and resource for future empirical investigation and theory-building. Before this however, attention turns to the nature of contingency theory and its relevance to civil society and the SRW.
In English the etymological roots of 'contingency' date from the mid-sixteenth century and derive from Medieval Latin for 'circumstance' (contingentia and contingere). It is used to denote dependence on certain (pre-)conditions. In conceptual terms, over recent decades the academic use of contingency theory in social and political science has ebbed and flowed. In various guises, it has been applied across disciplines and contexts -from management studies and organizational bureaucracy to postcolonialist analysis (Greenwood & Hinings, 1976). In part, this earlier use of contingency theory is the result of critiques of positivism and inductive generalizations about universalist norms (Bonacker, 2006). As Walzer (1990) notes, norms around meaning and understanding of the social world can only be validated when there is cognizance of local, contextual phenomena. This article adopts a similar position in relation to the civil society and the SRW. Ergo, the mere existence of particularistic factors (e.g. regular elections, civic activism, institutional capacity, and so on) does not, in and of itself mean that the SRW will be actualized. There is no inductive certainty involved. The straightforward co-presence of a core set of factors or preconditions applied across polities will not deliver universal results. Instead, contingency adds to civil society studies by underlining the complex interplay of contextual influences shaping social and political processes in a given civil society setting.
Here it should be noted that an important aspect of the contingency of SRW-namely, the conceptualization of what constitutes 'women's interests' in representative claimsmaking-is reflective of the fact that there is no universal set of 'women's interests' standing 'outside' of the representative process. In other words, there is no exhaustive SRW 'checklist' that political actors (such as elected representatives, women's movements, CSOs, and so on) may call upon when constructing their claims. This is in marked contrast to the position taken by earlier, largely essentialist and reductive analyses that were often based on the false premise of the existence of a discrete set of 'women's issues' (frequently defined as issues that mainly affect women, typically for biological reasons, such as reproductive rights, or for social reasons, such as childcare policy) (see Lovenduski & Karam, 2002).
In contrast, this article adopts the position that 'women's interests' crosscut all areas and aspects of public policy-and that they are actively constructed and negotiated during the representative process itself. This is captured in Lovenduski and Norris's (2003, p. 88) conceptualization of the SRW as 'a process of politicization'. In this, women's representation: can be treated as a number of steps in which (1) women are recognized as a social category, that is, the gender neutrality of politics is contested; (2) the inequalities of power between the sexes are acknowledged; and, (3) policies to increase the autonomy of women are made.
The model proposed in this article aligns with this proposition. It highlights the different structures and norms that feed into the representative process wherein 'women's interests' are actively constructed and negotiated through politicization.
In conceptual terms, the nexus between the SRW and contingency is expressed in the literature of critical realism (Bhaskar, 2008;Sayer, 2000). As Gerrits and Verweij (2013, p. 172) explain: reality is contingent. This means that any explanation [of social processes] is temporal in time and local in place. Since systems [such as those of political representation] are nested within their systemic environments, there is mutual influence between different systems.

They continue:
This property also implies that some mechanisms are in operation at given points of time, while others are not [ … this means] reality cannot be compressed without losing some of its aspects. In other words, while reduction or compression may be inevitable given the limits of human cognition and for practical research purposes, such a reduction or compression implies the loss of some of reality's properties such that any explanation is reductionist, i.e. an explanation can never fully contain the complexity it describes.
In this way, critical realists allude to the existence of innate causal powers (or what is dubbed 'natural necessity') in society (see Sayer, 1992Sayer, , 2000. Thus, in a given social situation, there may be sufficient conditions for a particular event or process to occur, but it happens only when the 'natural necessity' is triggered. To put it another way, the actualizing mechanism is contingent upon the alignment or co-presence of social objects with causal powers. Crucially, this has both temporal and spatial components. Again, this is relevant to civil society studies for the following reason: it underlines that any attempt to understand social processes in a given civil society setting needs to be cognizant of how the 'social space for action' is at once shaped by the historical development of the polity and also the areal socio-economic qualities specific to different geographical localities (inter alia, patterns and processes of social capital, the existence of inequalities in class and/ or wealth-and so on).
A key question here is what makes factors 'contingent factors'-as opposed to simply 'factors'? Jones and Hanham (1995, p. 188, emphasis added) offer an explanation: contingency is conceived not as a residual category, that is, as a new linguistic veil for particularity, but as an intervention in a process caused by context-dependent differences within which that process is embedded. Contingencies interrupt the operation of processes, thereby producing different empirical outcomes in different contexts.
In ontological terms, this has powerful implications for the study of civil society and the SRW. It means empirical work requires cognizance of a full range of contingent factors influencing and shaping CSOs' claims-making. 1 In other words, the particular way that the SRW operates in a given civil society context is intimately shaped by the alignment or co-presence of contextual factors from the prevailing political, socioeconomic, and organizational spheres (Figure 1). Given the myriad of variables at play (inter alia, democratic/administrative histories of polities, mobilizing structures, resources, skills and human capital, action repertoires, and so on; (see ensuing discussion)), these are unlikely to be identical in any two contexts. Rather, the way that the SRW plays out in a given context is contingent on local specificities. The challenge for social research is to systematically identify and acknowledge the ways that these shape the patterns and processes of substantive representation. It is the purpose of this article to address this challenge and posit a framework for analysis.

Analytical framework: contingent factors shaping CSOs' promotion of the SRW
In this section, we outline this study's analytical framework of CSOs' promotion of the SRW. Its grounding in extant studies and accompanying explanation of the constituent factors follows this initial overview.
The framework shows how the SRW in civil society is based on a staged processnamely CSOs' translation of motives ('collective incentives for political engagement') into action ('CSO action repertoires') in order to advance the SRW through claimsmaking. In turn, grounded theory from a survey of existing studies confirms how this is shaped by three overlapping contextual spheres: political, socioeconomic, and organizational ( Figure 1). As Table 1 summarizes, each of the collective incentives may be affected by-or be contingent upon-one, two, or all three of the spheres (the letter codes: O = Organizational, P = Political, S = Socioeconomic in-the central column in Table 1 headed 'Sphere' denote which spheres influence each respective factor). In a similar vein, the different components of action repertoires are also mediated by some or all of the spheres.
Attention now turns to consider the various contingent factors emerging from the grounded theory as listed in Table 1, beginning first with the collective incentives for  Figure 1). political engagement (labelled 1 a-f). This is followed by those shaping CSOs' action repertoires (similarly, labelled 2 a-f).

Contingent factors influencing the collective incentives for political engagement
Existing work points to how these foundational motives shape CSOs' propensity to advance the SRW. As Mazur (2013, p. 8, emphasis added) underlines in the context of multi-level governance: 'a more bottom-up approach to policy implementation that focuses on democratic processes implies the analysis of … feminist policy issues [needs to be cognizant of the] incentives for the development of feminist policy'. This assertion is grounded in a raft of empirical work such as that by Korolczuk (2014, p. 952) who notes the need to examine 'the specific solutions supposed to strengthen NGOs and give activists incentives to work on behalf of the common good [and] how they affect civil society actors' (see also Salmenniemi, 2005, p. 739).
In social theory terms, such incentives are captured by the collective interest model (Finkel, Muller, & Opp, 1989, p. 39), which outlines how group incentives for participation to challenge the prevailing political agenda fall into three categories, each of which relates directly to the SRW: (1) 'high levels of discontent with the current provision of public goods by the government or regime', (2) the belief 'that collective efforts can be successful in providing desired public goods'; and crucially for the present purposes, (3) the belief that CSOs' 'own participation will enhance the likelihood of the collective effort's success'. We now turn to the contingent factors.
The prevailing model of democracy, party/electoral politics Claims-making is contingent upon the opportunities afforded by the macro-political environment, including the prevailing model of democracy and party/electoral politics (see e.g. McAdam, McCarthy, & Zald, 1996, pp. 23-24). Key related questions include: does a universal franchise operate? (i.e. are women afforded a vote?) and what is the level of women's descriptive representation in parliament? (in other words, how many women parliamentarians are there?) Another key question relates to governing party turnover and associated 'cycles of contention'. The work of Tarrow and Tilly (2001) explains how these operate in regular liberal democracies: new CSO interest representation tactics arise in the context of peaks or cycles of protest activity (or 'cycles of contention'). Foremost of these are electoral cycles. However, in polities lacking governing party turnover (single party dominant systems), the catalysing effect is lost and CSOs may become party-institutionalized and de-radicalized. Instead, they may be forced to engage with the same party in office over extended periods. This may lead to a stasis in claimsmaking and stagnation setting in. Such an example illustrates the contingent effect of the macro-political environment on CSO motives for SRW claims-making-in this case it is contingent upon governing party turnover and accompanying cycles of contention.
By way of further illustration, an established literature outlines how elected representatives may seek to determine the timing of elections to seek electoral advantage (Meredith, 2009). What has received far less attention however is the impact of election timing on groups in civil society and the SRW (see Schlozman & Tierney, 1986). Some studies (e.g. Maundeni, 2005) point to heightened civil society lobbying during the periods when party manifestos are drafted, the rationale being that political opportunity structures are more conducive to claims-making when parties are actively seeking policy proposals to include in their election programmes. This is supported by a growing trend of CSOs launching their own manifestos with the explicit purpose of claims-making and shaping party programmes (Ntseane, 2005). A contrasting proposition comes from Anzia (2011), who argues that off-cycle election timing and subordinate ballots create a strategic opportunity for organized interest groups. In the case of the SRW, further comparative work is needed to explore which of the foregoing scenarios applies in order to better understand the contingent impact of election timing on claims-making. Again, the underlying point from this illustration is that scholarly work on civil society and women's representation needs to be alive to the fact that the motivations of CSOs are contingent upon wider macro-political aspects, in this case the timing of elections.
Another illustration of how CSOs' incentives for promoting the SRW are contingent in nature is provided by the effect of 'veto players'. These are individuals who, by virtue of their office, may block exogenous claims-making on an issue (for a discussion see Tsebelis, 1995). In the present case, they are parliamentarians and party figures who are unreceptive to SRW claims (for reasons including political ideology, opposition to interventionist policies, personal prejudice, and discriminatory attitudes). The pathology that they present to civil society political engagement around the SRW (and other matters) is heightened in polities characterized by single party dominance and low governing party turnover. In such contexts, veto players have enduring influence (e.g. they are less likely to lose ministerial positions as a result of their party being voted out of office)-and may exclude SRW claims over successive electoral cycles.
In terms of the grounded theory underpinning the present analytical framework, the contingent effects of the macro-political environment pervade the wider literature. A specific example is provided by Phillips (2005). Reflecting upon gender, civil society, and social activism in post-Soviet Ukraine, she concludes: 'these processes, and the contradictions they entail, challenge us to acknowledge the contradictory personal transformations that striving for social and political change can engender, while remaining attuned to the structures of power that constrain agency'. In a similar fashion, Moon (2002, p. 474) concludes her study of civil society and the women's movement in South Korea: 'this line of feminist analysis can serve as a conceptual tool to examine the impact of political democratization on women and, in return, women's ability to shape this process in industrialized Asian societies'.

Trust
Trust is a key contingent factor shaping CSOs' collective incentives for political engagement. It is a core component of bridging social capital and a predictor of policy engagement (Mishler & Rose, 2001). In turn, it shapes CSOs' collective incentives for political engagement. The underlying logic is if members of CSOs feel that politicians and legislative proceedings (e.g. petitions committees, government inquiries, policy consultations, and so on) are trustworthy, they may be selected as a means to advance claims on those in power (Fennema & Tillie, 1999).
In contrast, if these mechanisms are deemed untrustworthy, alternative means may be employed (e.g. direct action, boycotts, demonstrations). The level of trust is founded in individuals' perceptions of a range of considerations, including party politics in legislative settings (Dunn, 2011, p. 396), as well as the executive party's past record and political performance (Price & Romantan, 2004). As Lühiste (2006, p. 493) puts it, the latter can be described as 'the regime's capacity to produce the so-called procedural goods and desired outputs such as fair treatment of its citizens, protection of civil liberties, and transparent and effective administration'. Grounded illustration of the role of trust as a contingent factor can, for example, be seen in the work of Dris-Aït-Hamadouche (2007, p. 127) in relation to CSOs and promotion of the SRW in the Maghreb: 'secular organizations criticize Islamist parties because they mistrust the Islamist women's motives. In their view, the involvement in women's commissions represents a strategy of infiltration and power grabbing'.

Governance structures
Prominent examples of the contingent influence of governance structures on the SRW include the instruments of state feminism, that is, institutional mechanisms built into representative structures and governance practices designed to advance women's representation. Such measures include cross-party gender equality committees in legislative settings, state funding for women's policy-networks, and inter-sectoral partnership arrangements. The latter are formal institutional structures linking civil society and state designed to facilitate participative policy-making. The significance of such governance structures is that they shape system openness, conduciveness to SRW claimsmaking and the prevailing political opportunity structures. The presence/ absence of such factors (dis-)incentivizes CSO to engage in politics and policy-making to advance women's interests (cf. Chaney, 2008a). Their absence makes it hard for such concerns to be heard amidst competing claims in liberal democracies. The salience of governance as a contingent factor in the wider civil society literature is illustrated by Jahanshahrad's (2012, p. 234) work on civil society and the women's movement in Iran where she makes the case for revised governance practices and instruments of state feminism to 'engender' (or bring gender into) policy-making: civil society has been the realm of male-dominated policy-making processes and also a realm of disintegration and gender-based inequalities. However, it is important to identify the possibilities that civil society can provide for the emergence of public spheres within which subordinated and marginalized social groups such as women can articulate their concerns and develop new ideas.

Economy
Economic considerations affecting the SRW include the available resources for gendered policy reform. The presence of a budget surplus and economic growth-as well as the existence of government grants and funding to advance gender equality-all shape the incentives for CSO engagement. Conversely, gendered inequalities, periods of austerity, and the absence of state funding to support the SRW may dis-incentivize policy engagement by CSOs. As a contingent factor, economic considerations pervade the civil society literature on women's representation, a prominent example being Sloat's (2005, p. 437) work on the growth of women's NGOs in Central and Eastern Europe. Here she illustrates the centrality of the economy as a contingent influence: Lithuania provides a typical example. Most women's organizations were formed between 1992 and 1996 after the country entered a phase of democratic consolidation and achieved macroeconomic stability. They include the revival of some historical women's self-help associations and the establishment of women's clubs, societies and study centres. The situation in the Czech Republic is similar … Cultural attitudes/prevailing gender relations Axiomatically, the prevailing state of gender relations determines whether there is a need to tackle gender oppression or whether a post-feminist, gender-equal society pertains. Cultural practices and social attitudes are further contingent factors here (Inglehart & Norris, 2001. As George's (2007, p. 679) insightful account underlines, there is a need to 'locate and situate gender mainstreaming [-and the SRW more generally,] in the culturally-specific contexts in which it is practiced to capture the complex realities in which gender policies are implemented and women are positioned to effect change'. Thus, for example, analysis of the SRW needs to consider how it is shaped by factors such as religion, ethnicity, caste, and local traditions that affect social attitudes and gender norms regarding women's role across, domestic, economic, and public spheres. In this regard, key studies underline how CSOs' effectiveness is shaped by the degree of (dis)connect with local traditions (cf. Nezhina & Ibrayeva, 2013). Specific examples include Ferree's (2003) work on civil society and feminist activism on abortion policy in the USA and Germany and Jahanshahrad's study of Iran where she concludes (2012, p. 238): my own analysis suggests that the political structure in a nation interrelates with the culture of that society. While the political system plays a crucial role in shaping the political and public culture of society, these cultures have a decisive impact … [androcentric values have] prevented the legal establishment of independent political and social organizations … genuine civil society in Iran has faced fundamental barriers.

Embeddedness of gender equality laws
This contingent factor refers to the existence of mechanisms of redress for sex discrimination/gender oppression. The prevailing legal framework in a given civil society context determines whether, in advancing the SRW, CSOs are building on earlier interventions that have secured a degree of progress-or whether they are working from a 'zerobase' in polities characterized by ongoing gender inequality. Allied to this, the monitoring and enforcement of anti-discrimination law is a further contingent factor. In other words, scholarly work needs to consider the extent to which government is bound by 'fourth generation' or 'positive' equality duties that require the promotion of gender equality-including measures that actively facilitate the participation of groups with 'protected characteristics'-(including women) in politics and policy-making (cf. Dobrowolsky & Hart, 2003;Fredman, 2001Fredman, , 2008. Moreover, there needs to be cognizance as to whether governments have ratified international human rights instruments and treaties such as the UN Declaration on Human Rights and, the Convention on the Elimination of all forms of Discrimination against Women. Again, the presence of gender equality/anti-discrimination law as a contingent factor pervades the civil society studies referring to women's representation. For example, it is illustrated by Salmenniemi's (2005, p. 742) work on feminism, civil society, and citizenship in post-Soviet Russia. She concludes: many women voiced discontent and would like to change the situation [prevailing gender oppression]. This, in turn, can serve as a catalyst for politicizing the private sphere and thus for renegotiating the terms of citizenship. However, not all female interviewees perceived themselves as participants in this process. [Yet … ] young women in the human rights organization considered the state and the legal system responsible for resolving this question [i.e. enabling them by providing them with a platform or 'voice' for protest].
Attention now shifts from motives-to the contingent factors associated with CSOs' different modes of advancing the SRW-or 'action repertoires'.

CSO action repertoires
Once the decision is made to advance SRW claims on those in power, CSOs have to choose the method or means by which to prosecute their policy demands. The resulting modes of engagement constitute CSOs' action repertoires (alternatively, 'repertoires of contention') (see Tilly, 1978, p. 42). The full gamut of claims-making methods open to CSOs spans a broad range including: violent protest and civil disobedience at one end of the spectrum through to electronic media campaigns -and on to more regulated and bureaucratized means such as policy lobbying and policy consultation work. Different actions will have different levels of success in advancing the SRW in policy and politics. Crucially, as the following discussion reveals, CSOs' choice of action repertoire is contingent on the broader socioeconomic, political, and organizational context.
CSOs' human capital/organizational culture and practices 'New' or neo-institutionalism underlines the importance of rules and norms in shaping the way that institutions-including CSOs-operate (Hall & Taylor, 1996). A strand of this literature, feminist institutionalism, explores the manner in which political institutions may reflect and mediate gendered patterns of power (Wiegman, 1999). As Olsen (2009, p. 9) notes, on balance, institutional change tends to be internally-rather than externally-driven. Applied to CSOs and the study of the SRW, this suggests the need for cognizance of 'the internal success criteria, structures, procedures, rules, practices, career structures, socialization patterns, styles of thought and interpretive traditions, and resources of the [organization]' (Olsen, 2009, p. 9). Because organizational culture comprising norms and values sustains, reflects and variously reinforces or challenges gender relations in a given institution (Gherardi, 1995) it is a contingent factor that shapes not only the extent to which CSOs act to promote the SRW both internally and externally -but the manner in which they do so.
Organizational culture and practices are sustained by human agency (Bovey & Hede, 2001, p. 534). Thus the staffing and membership of CSOs also matters to the promotion of the SRW. Here a contested part of the extant literature suggests that probabilistically, in legislative settings at least, women are more likely to advance the SRW than men (Childs & Withey, 2004). Thus, in the present case, contingent factors shaping the SRW include the gender balance in the staffing and membership of CSOs. In turn, this relates to debates over the relative influence of 'critical mass' (or, the overall number of women in an organization) versus 'critical actors' (in other words, key individuals that have a disproportionately strong influence in the promotion of the SRW compared to their peers) (cf. Childs & Krook, 2009). Notably, vertical gender segregation issues are also influential here. In other words, the way that the SRW is promoted is not only shaped by the number of women in the organization and presence of critical actors, as Prouteau and Tabariés's (2010) insightful account reveals, it is also influenced by the extent to which women hold positions of power and influence in CSOs (see also Blackburn, Browne, Brooks, & Jarman, 2002). Allied to this, a further contingent organizational factor is the internal governance practices and agenda-setting within CSOs. As Egeberg (2003) explains, bureaucratic structures and agenda-setting procedures matter to organizational democracy and claims of representativeness and accountability. When applied to the SRW a key issue here is whether claims are determined by 'grassroots' consultation with CSO members or decided by managers on an 'executive'-but not necessarily democratic-basis.
Allied to the foregoing is organizational leadership. It encompasses issues such as vision, boldness, and the degree of radicalism characterizing CSO claims-making. These factors provide organizational resources that are crucial to the translation of conviction into action. Existing work points to complexity in understanding such links in relation to leadership and gendered outcomes. Contingency is introduced by the prevailing models of leadership in CSOs (inter alia, transformational, distributive, and charismatic). This influences SRW claims-making by shaping a diverse range of factors, including functional competence, strategizing, and workforce/ membership unity (see Engen, Leeden, & Willemsen, 2001). The presence of human capital and organizational culture as a contingent factor is widely acknowledged in the literature. It is illustrated by Von Doepp's study (2002, p. 276) of civil society, local churches, and women's empowerment in rural Malawi. He concludes: more than local Presbyterian churches in the area, the local Catholic Church offered women an organizational environment wherein they could acquire important skills and take on unique public roles. This contributed to their being active and vocal in local politics. What is especially notable about these findings is that the organizations that conformed most to the liberal ideal of civil society (i.e. the Presbyterian churches) proved least effective in encouraging public engagement among marginalised citizens … issues that, together, expose problems in the liberal understandings of civil society.

Capacity/financial capital/resources of CSOs
Resources and institutional capacity are additional contingent factors shaping CSOs' claimsmaking-as set out in resource mobilization theory (McCarthy & Zald, 2002). In essence, they reflect the straightforward reality that dissent, activism and grievances alone will not deliver desired social change. Instead, in addition to political influence, CSOs' claims-making can, in part, usefully be seen as a struggle over resources (in the case of the SRW this is evident in calls for governments to adopt gender budgeting and redistributive social policies). Axiomatically, a CSO's access to resources also determines the organization's ability to function effectively across a range of domains and activities. The centrality of organizational capacity as a contingent factor is evident not only in extant empirical studies-such as Fuchs and Payer's (2007) work on the capacity of Central and East European interest groups to participate in EU governance-but also in the international policy framework. For example, the UN's strategy 'The Premise and Promise of UN Women's Partnerships with Civil Society' (UN, 2013, p. 3) gives as a core priority the goal of 'strengthen[ing] the capacity of CSOs to more effectively advocate gender equality and women's empowerment to be central to the new development agenda'.
Networking practices/civic traditions Networking practices and civic traditions are also contingent factors. Notably, these were given prominent attention in Putnam, Leonardi, and Nanetti's (1993) classic work on governance and social capital in Italy. They cover a range of matters including the history of civic activism in a given polity (Skocpol, 1997). In turn, networking practices and civic traditions link to constitutional and political history, for example, whether a polity has a tradition of authoritarian or libertarian rule. This set of contingent factors also relates to patterns and processes of associative life and whether a society is closely networked and engaged in public decision-making, or atomistic and detached. Organizational structures that sustain collective action in civil society through coordination and collaboration between CSOs exert a contingent influence on the SRW and, in turn, raise a number of issues and challenges (cf. Deo, 2007;McAdam et al., 1996, p. 13). For example, isolated 'standalone' organizations may prove weaker and thus easier for political elites to sideline or defeat compared to co-ordinated action from multiple CSOs.
Conversely, the involvement of greater numbers of CSOs may raise coordination issues and increase the potential for disagreement and division (see e.g. Banazak, 2010). Such challenges are captured in Sabatier and Jenkins-Smith's (1993) classic conception of 'advocacy coalitions', which effectively regard the policy process as a competition between rival groups (or coalitions) of claims-makers, each seeking to advance their beliefs about policy problems and solutions-such as women's representation. The presence of civic traditions of activism in the current study's analytical framework is grounded in empirical studies such as that by Korolczuk (2014, p. 755) on feminist mobilization and the development civil society. Here the author concludes: 'there is a relation between the low level of civic activism in Poland and the way NGOs function, and that the growing bureaucratization and professionalization of the latter discourages people from joining organizations and groups'.
Institutional configuration of the state The types of state 'machinery' present in a polity will determine the means by which CSOs engage those in power (e.g. using virtual techniques and web-based engagement-or more traditional forms of lobbying and claims-making, including protest marches). Thus the burgeoning neo-institutional literature (cf. Peters, 2012) underlines how the institutions of the state constitute 'a relatively enduring collection of rules and organized practices, embedded in structures of meaning' (March & Olsen, 1984, p. 4). At the most fundamental level lies constitutionalism, or the idea that governments must act in accordance with a known constitution (Tully, 1995). This is particularly salient in the era of multi-level governance (Haussman & Sauer, 2007) and shapes civil society action repertoires on the SRW through the basic need to align claims-making with the powers of different tiers of government.
Beyond constitutionalism, institutional configuration has further major implications for issues of accountability and SRW claims-making (Olsen, 2013). In particular, as Lowndes and Wilson (2001) emphasize, 'institutional filters' (or the openness of the design of political institutions) act as a key explanatory variable as to why participation rates in government institutions vary between different social groups in civil society, and notably along gender lines. They also determine the openness and accessibility of institutions to exogenous claims-making and the way that policy elites process and respond to such claims.
As Squires (2008, pp. 187-188) notes, the political opportunity structures of a given political system will 'privilege particular conceptions of group relations over others'. This is because, as Lowri's (1971) classic work underlines, 'areas of policy or government activity constitute real arenas of power. Each arena tends to develop its own characteristic political structure, political process, elites and group relations ' (1971, pp. 689-690). This condition has interesting implications for CSOs and the SRW. It means that claimsmaking will be shaped in a contingent fashion with variation between policy areas and, for example, government ministries, depending on the singular political dynamic and power relation applying. Here, the rubric of feminist institutionalism (Haussman & Sauer, 2007) provides an interrogatory framework to explore exactly how state structures contingently facilitate or frustrate SRW claims-making. In the former regard, it may be aided by the presence of state feminism, or 'government structures that are formally charged with furthering women's status and rights' (Stetson & Mazur, 1995, p. 2). The present survey of the empirical literature reveals a diverse range of institutional aspects of state configuration that exert a contingent effect upon the SRW. Strategic partnerships between government and CSOs are a leading example, as Halsaa's (1999) work on strategic partnerships for women's policies in Norway and Holli and Kantola's (2005) study in Finland attest.
Organizational resilience and psychological capital CSOs' SRW claims-making will also be shaped by organizational resilience and psychological capital. As Burnard and Bhamra (2011) explain, organizational effectiveness over time is shaped by the ability to withstand external shocks as well as deal with internal problems across a range of domains (including staffing, resources, management, leadership, and strategy). In the present case, claims-making by CSOs is contingent on their ability to endure opposition and the rejection of earlier attempts to advance the SRW claims on those in power (see Edwards & McCarthy, 2004). It also includes resilience in the face of resource shortages (e.g. expenditure on failed rounds of claims-making, funding cuts, and so on) and skills loss (e.g. due to key personnel leaving an organization). In turn, such resilience is shaped by employees' and members' psychological capital, that is, positive psychological resources-including hope, efficacy, and optimism (Luthans, Avey, Avolio, Norman, & Combs, 2006). Conversely, a lack of resilience may result in a loop of negative feedback introducing a range of organizational pathologies including demoralization (Limnios, Alexandra, Mazzarol, Ghadouani, & Schilizzi, 2014).
Allied to the foregoing, civil society claims-making on the SRW is further influenced by organizational learning and adaptation, because the extent to which CSOs can successfully adapt their practices in the face of earlier setbacks and mistakes affects future effectiveness (Sabatier & Jenkins-Smith, 1999). As March (1991, p. 71) notes: a central concern of studies of adaptive processes is the relation between the exploration of new possibilities and the exploitation of old certainties. Exploration includes things captured by terms such as search, variation, risk taking, experimentation, play, flexibility, discovery, innovation. Exploitation includes such things as refinement, choice, production, efficiency, selection, implementation, execution.
Empirical grounding for organizational resilience and psychological capital as a contingent factor comes from a range of studies including Twigg's (2009) work on gender and community organizations in disaster-relief efforts in Bangladesh and Nepal. In this example, CSOs faced funding uncertainty, rivalry from international agencies, and the loss of key personnel and institutional expertise.

Norms
Feminist institutionalist research has explored the contingent relationships between feminists in political institutions, systemic factors related to the working of the polity (including electoral and party systems) and norms (such as norms of equality, rights, and representation)-and whether they facilitate or frustrate women's substantive representation (see e.g. Childs & Lovenduski, 2012). From an international perspective Kantola and Squires (2012, p. 389) also point to how: international norms and transnational networks are now key to the policy-making process … These norms, which are embedded in international treaties, declarations and policy recommendations of international organizations, have increasingly included the status of women as a significant focus … [they have had the effect of] creating international gender equality norms that have been widely diffused across the globe. Kantola and Squires (2012, p. 389) proceed to underline the influence that norms have on CSOs by concluding: 'this, rather than local social movement activism and lobbying, has generally placed the pursuit of gender equality onto state policy agendas'.
In a similar vein, Von Doepp (2002, p. 278) and Waylen (2008, p. 123) are unequivocal in underlining the contingent nature of norms in shaping CSOs' action repertoires. The former states: 'both primary social relations and norms may be the fundamental constraints to political participation by the marginalised'. Whilst the latter adds that CSOs are no different to other 'social institutions which embody social norms and practices, [and] are imbued with power relations that include a gender dimension'. Allied to this, Marx et al. (2002, p. 312) are clear on the pivotal way norms shape CSOs' discourse and policy framing: 'the norms and practices governing policy discourse privilege certain forms of representation over others, and thus selectively disempower certain categories of speakers'. In this way, action repertoires are contingent upon prevailing social norms, or 'templates on which change is measured, both with respect to the beliefs of different coalitions and the actual content of public policy' (Sabatier, 1993, p. 55).
In turn, prevailing norms shape CSOs' discursive politics in order to advance the SRW and secure gender transformation. Discursive politics are thus a crucial dimension of power, with norms shaping the way policy actors engage in conceptual disputes, influenced by the distribution of material and institutional power. Unsurprisingly, the salience of norms to CSOs' action repertoires is underlined in a broad range of studies, typified by Sloat's (2005, p. 243) work on eastern Europe where, for example, she concludes: in Hungary, the women's civic sector suffers from external weaknesses (including undemocratic institutional mechanisms and lack of financial support) and discord between organizations (stemming from a lack of lobbying experience, knowledge of democratic norms and understanding of the importance of collective representation).
In a further example, Chattier's (2015, p. 177) work on CSO mobilization in Fiji underlines the contingent effects of prevailing norms. She concludes: notwithstanding: women's movements and civil society activism becoming more astute to concerns of gender equality and lobbying for women's political participation … patriarchy is still a major force hindering women's political advancement in Fiji. A combination of cultural stereotyping and persistent gendered norms contribute to the masculinization of the political realm and eulogise women's role in the private sphere.

Conclusion
The foregoing systematic analysis makes an original contribution to civil society studies by exploring civil society's role as a political space integral to the SRW. It offers a corrective to earlier, reductive accounts of women's representation that lack explanatory power because they portray civil society as an homogenous, largely unquestioned arena outside the state, economy, and familial spheres. It also responds to the fact that much of the extant political science literature on women's representation has focused exclusively on legislative settings. In response, this study's analytical framework is grounded in a raft of empirical studies and underlines that effective representation cannot be divorced from context. Thus its key contribution is in highlighting the importance of civil society for future study of substantive representation and in affirming earlier, under-developed assertions in the political science literature that women's representation outside legislative settings is a contingent process shaped by three (non-discrete) spheres: the political, socioeconomic, and organizational.
This conceptual synthesis also points to a future research agenda that includes qualitative work on: (a) management and leadership processes in CSOs and the way in which they mediate contingent factors when SRW claims are made; (b) the experience of 'grassroots' members of CSOs and the way in which CSOs' organizational practices engage them in shaping the SRW claims made in their names; (c) responsiveness and intersectionality -specifically, the discursive process of claims-making and whether this facilitates or frustrates claims-making that reflects women's multiple and simultaneous identities (i.e. in terms of ethnicity, age, (dis-)ability, and so on); (d) the particular influence of electoral cycles on the timing and nature of SRW claims-making by CSOs; and (e) how contingency affects claims-making across different policy areas and between different government ministries. Application of this article's analytical framework to the foregoing agenda will advance civil society scholarship through conceptual synthesis, thereby producing more sophisticated and nuanced research that is cognizant of the co-presence of contingent factors that offer immanent explanatory power of how civil society acts as a gendered political space for advancing the SRW. Note 1. 'Claims-making' is a social constructivist perspective on representation that pays particular attention to discourse and the language used in seeking to represent different groups and constituencies (see e.g. Loseke, 2011). It is used more broadly in this article to signify a broad range of action repertoires used to advance demands-or claims-on those in power in order to advance the representation of women.